Chlamydia trachomatis, tubal disease and the incidence of symptomatic and asymptomatic infection following hysterosalpingography.

The risk of causing or reactivating pelvic infection by hysterosalpingography (HSG) was assessed in 118 infertile women. Serological evidence of Chlamydia trachomatis infection was sought before, and 10 days and 4 weeks after HSG, using the single-antigen whole-inclusion immunofluorescence (WIF) test for species-specific antibody and the complement fixation test (CFT) for group antibody. Chlamydia antigen was detected using an ELISA. There was a close correlation between the finding of occlusive tubal damage and serum antibodies to C.trachomatis detected by both tests (P less than 0.001). Of 60 patients with WIF antibody titres greater than or equal to 1/64, 65% had evident tubal damage compared with only 16% of patients without a raised titre. Clinically suspected sepsis after HSG occurred in 4% of cases both with and without antibodies to C.trachomatis but only in those with tubal damage (10%). There was no serological evidence of C.trachomatis involvement in symptomatic or asymptomatic patients and anaerobic pathogens were isolated more commonly than C.trachomatis in those symptomatic cases. Isolation of C.trachomatis in two cases from the endocervix and also from the endometrium in one of these only after HSG was compatible with reactivation but not reinfection. Mycobacterium tuberculosis was isolated in two out of three cases with unsuspected tubal disease without antibodies to C.trachomatis. The risk of infection from HSG appears to be confined to patients with existing tubal damage, and in those unexpected cases without serological evidence of previous chlamydial infection, tuberculosis seems to be a likely cause.